You are on page 1of 5

Proc. Nail. Acad. Sci.

Vol. 89, pp. 7203-7207, August 1992

Dermorphin-related peptides from the skin of Phyllomedusa bicolor
and their amidated analogs activate two ,u opioid receptor subtypes
that modulate antinociception and catalepsy in the rat
([Lys7]dennorphin/[Trp4,Asn7]dermorphin/[Trp4,Asnsldermorphin-(1-5)/guinea pig ileum/mouse vas deferens)
*Institute of Pharmacology, University La Sapienza, 00185 Rome, Italy; and *Institute of Neurobiology, National Research Council, 00185 Rome, Italy
Contributed by Vittorio Erspamer, April 20, 1992

ABSTRACT Three naturally occurring dermorphin-like chemistry on a Biolynx automated peptide synthesizer (Phar-
peptides from the skin of the frog PhyUomedusa bicolor, the macia Biochrom, Cambridge, U.K.). The products were
related carboxyl-terminal amides, and some substituted analogs purified by preparative HPLC on a Vydac reverse-phase
were synthesized, their binding profiles to opioid receptors were column (22 x 250 mm, 10-,um particle size, 300-A pore
determined, and their biological activities were studied in iso- diameter, C18 packing; The Separations Group) with a linear
lated organ preparations and intact animals. The opioid binding gradient from 1o to 90% of solvent B (5 mM trifluoroacetic
profile revealed a very high selectivity of these peptides for IA acid in acetonitrile) in solvent A (5 mM trifluoroacetic acid in
sites and suggested the existence of two receptor subtypes, of water) over 60 min. After hydrolysis in 6 M HCO (24 hr, 1100C,
high and low affinity. The peptides tested acted as potent IA in vacuo), amino acid analyses gave ratios consistent with the
opioid agonists on isolated organ preparations. They were desired structures. The expected amino acid sequences were
several times more active in inhibiting electrically evoked con- confirmed by automated Edman degradation performed with
tractions in guinea pig ileum than in mouse vas deferens. When an Applied Biosystems 470A protein sequencer.
injected into the lateral brain ventricle or peritoneum of rats, the Binding Assays. Binding of the peptides to p., 8, and K
high-affinity-site-preferring ligand, [Lys7-NH21dermorphin, be- opioid sites was assayed in crude membrane preparations (6)
haved as a potent analgesic agent. By contrast, the low-affinity- from rat (p., 8) and guinea pig (K) brain at pH 7.4 in 50 mM
site-preferring ligand, [Trp4,Asn7-NH2]dermorphin, produced Tris HCl buffer. Each assay mixture contained, in a final
a weak antinociception but an intense catalepsy. volume of 2 ml, the membrane preparation (0.8-1.0 mg of
membrane proteins), protease inhibitors (100 puM phenyl-
Dermorphin is an opioid peptide (see peptide 1 in Table 1) methanesulfonyl fluoride, soybean trypsin inhibitor at 20
isolated from the skin of South American frogs belonging to ,ug/ml, and bestatin at 10 pug/ml) (Sigma), and the tritiated
the subfamily Phyllomedusinae (1). It has high affinity and ligand at the desired concentration with or without unlabeled
selectivity for p. opioid receptors and produces analgesia in ligand. The p. binding site was selectively labeled with [tyrosyl-
laboratory animals and humans (2-4). By means of a cDNA 3,5-3H][DAla2,Me-Phe4,Gly-ol5]enkephalin ([3H]DAGO, Am-
library prepared from the skin of Phyllomedusa bicolor, the ersham, UK) (0.5 nM); the 8 binding site with [3,5-3H]Tyr-
amino acid sequences of three dermorphin-like peptides were DAla-Phe-Asp-Val-Val-Gly-NH2, ([3H][DAla2]deltorphin I)
recently predicted (5). These dermorphin analogs have now (7) (0.3 nM); and the K binding site with (5a,7a,8f8)-(-)-N-
been isolated from methanol extracts of skin from this frog, methyl-N-[7-(1-pyrrolidinyl)-1-oxaspiro(4,5)dec-8-yl][3 ,4-
and their amino acid sequences have been confirmed (15). 3H]benzeneacetamide ([3H]U-69,593; DuPont/NEN) (1 nM).
Like dermorphin, these peptides contain a D-alanine as the Radioligands were checked for purity by reverse-phase HPLC
second amino acid and share the common amino-terminal (Vydac C18 column, 5-pum particle size, 300-A pore size, 2.1 x
sequence Tyr-DAla-Phe. The D-alanine present in the final 250 mm) and eluted according to the linear gradient described
products is encoded in the skin mRNA by a normal codon for above. Nonspecific binding was determined in the presence of
L-alanine. While other opioid sequences in the frog skin 50 pM naloxone (S.A.L.A.R.S., Como, Italy) for the p. and K
precursors are all followed by a glycine residue required for the systems or 5 ,uM naltrindole (Research Biochemicals, Natick,
formation of terminal amides, this is not the case for these MA) for the 8 system. After a 90-min incubation at 350C, the
three dermorphin-like peptides. Accordingly, the end prod- free ligand was separated from membrane-bound ligand by
ucts extracted from the skin were found to contain a free filtration over Millipore AP40 glass fiber filters (soaked in 0.1%
a-carboxyl group at the carboxyl terminus. In the previous bovine serum albumin incubation buffer for 1 hr) under re-
paper (5) the binding affinity and selectivity of these peptides duced pressure; the filters were then washed three times with
for p. opioid receptors were briefly reported. In this paper, we 5 ml of ice-cold buffer. The radioactivity was extracted in 10
present the binding profile to p., 8, and K opioid receptors and ml of Kontrogel (Kontron, Milan) and measured in a liquid
the biological activities of the Phyllomedusa bicolor dermor- scintillation counter (Betamatic V, Kontron). The percent of
phin-like peptides, their related carboxyl-terminal amides, and depletion of free radioligand by binding in the absence of a
a few substituted analogs. competing ligand (8) was 4% for [3H]DAGO and [3H]U-69,593
and 6% for [3H][DAlaIdeltorphin I. Competition curves were
MATERIALS AND METHODS determined in triplicate and were usually obtained over the
concentration range of 1 pM to 120 p.M. The inhibition
Synthesis of Peptides. The peptides were synthesized by
using fluorenylmethoxycarbonyl-polyamide active-ester
Abbreviations: DAGO, [DAla2,Me-Phe4,Gly-ol5]enkephalin;
U-69,593, (Sa,7a,8f3)-(-)-N-methyl-N-[7-(1-pyrrolidinyl)-1-
The publication costs of this article were defrayed in part by page charge oxaspiro(4,5)dec-8-yllbenzeneacetamide; [DAla2]deltorphin I, Tyr-
payment. This article must therefore be hereby marked "advertisement" DAla-Phe-Asp-Val-Val-Gly-NH2; i.c.v., intracerebroventricular.
in accordance with 18 U.S.C. §1734 solely to indicate this fact. tTo whom reprint requests should be addressed.
7204 Pharmacology: Negri et al. Proc. Nati. Acad. Sci. USA 89 (1992)
constants of the various nonradioactive peptides were calcu- g at the time of surgery were housed singly in 25 x 35 cm
lated by fitting the displacement curves with the nonlinear cages placed in a thermostatically controlled cabinet at an
regression program LIGAND (9), using one-site or two-site environmental temperature of 21'C. Under light diethyl ether
models. The results are given as mean SEM of at least six
± anesthesia, each rat was implanted surgically with a plastic
separate determinations. IC50 is the concentration of the guide cannula (Linca, Tel Aviv) screwed into a skull hole
competing ligand at which specific binding of labeled ligand is drilled over the left lateral cerebral ventricle (AP, -0.5 mm
reduced by half. Kd and Ki are the equilibrium dissociation and L, +1.8 mm relative to bregma; V, -1.0 mm relative to
constants for the radioligand and inhibitor respectively, 1H is the skull surface). The cannula was secured to the bone with
the slope of the log-logit plot of competition data, and B. is dental cement and the rat was returned to its cage and allowed
the maximum binding capacity of the radioligand. Because it to recover from surgery for at least a week. Peptides dis-
is impractical to measure concentrations of unbound high- solved in saline were delivered to the lateral ventricle through
affinity unlabeled competing ligands, whenever an observed the plastic cannula, with a Hamilton microliter syringe. All
IC5o is less than the radioligand Kd, the computed value of K; i.c.v. injections were made in a volume of 5 dl.
may be too large by an indeterminate factor (8). Test of Antnociception. Antinociceptive responses to pep-
Pharmacological Assays in Isolated Tissues. Preparations of tide administration were determined in rats by the tail-flick
the myenteric plexus-longitudinal muscle obtained from the test (12). The latency to the first sign of a rapid tail-flick was
small intestine of male guinea pigs (400-500 g) and prepara- taken as a measure of the nociceptive response. Before
tions of mouse vas deferens were used for field stimulation peptide administration, each rat was tested and the latency to
with bipolar rectangular pulses of supramaximal voltage as flick was recorded (control latency, CL). Animals not flicking
described by Gyang and Kosterlitz (2). Synthetic reference their tails within 3 sec were discarded. The test was repeated
peptides were dermorphin (Farmitalia) for guinea pig ileum, every 15 min during the first hour after i.c.v. administration
and [DAla2jdeltorphin 1 (7) for mouse vas deferens prepara- of peptides and thereafter every 30 min for a total period of
tions. Assays were performed in the presence or absence of 4 hr. The latency to flick of peptide-injected rats was defined
naloxone and the 8 opioid receptor antagonist naltrindole as the test latency (TL). To avoid tissue damage, rats with a
(10). The results were expressed as the ICso values obtained test latency of more than 10 sec (cutoff time), were removed
from concentration-response curves. When peptides were from the nociceptive stimulus and assigned a maximal anti-
assayed, [DAWa2]deltorphin I and dermorphin were used as nociceptive score of 100%/. For drawing dose- and time-
internal standards with mouse vas deferens and guinea pig response curves, antinociception (A) was expressed as A =
ileum preparations, respectively. Because preparations var- 100 x (TL - CL)/(10 - CL). To determine if antinociceptive
ied considerably in sensitivity, the results obtained for the effects were the result of activity at opioid receptors, nalox-
various peptides were normalized as described by Kosterlitz one (0.1 mg/kg, s.c.) was administered 15 min before i.c.v.
et al. (11). The ICso value obtained in an individual assay was injection ofthe peptides and antinociception was measured as
multiplied by the following fraction: mean IC50 value of the described above. Each peptide dose was tested in a 10-rat
standard determined for all the assays IC50 value of the
+ group. The AD50 of each peptide was defined as the dose that
standard found in the individual assay. produced an antinociceptive response in 50% of the animals
Test of Peptide Degradation. The integrity of the radioli- tested. An antinociceptive response was considered to occur
gands after 90-min contact with the membranes was exam- when an individual animal displayed a TL value greater than
ined by HPLC as described above; no degradation was its CL value plus three SDs of the control mean CL of all
observed. Degradation of the peptides was also determined animals in the group.
by incubating them with rat brain homogenates in 50 mM Test of Catapsy. Catalepsy was evaluated by placing both
Tris HCl, pH 7.4, at a concentration of 100 .ug of peptide per front limbs of the rat over a 10-cm-high horizontal bar and
g of tissue, in the presence or absence of the protease measuring the time the animal maintained that posture. Rats
inhibitor cocktail used in the binding studies. After incuba- remaining more than 1 min on the bar were defined as
tion, peptides were extracted with methanol and the recovery cataleptic (positive bar test). The cataleptic effect of each
was determined by pharmacological assay on mouse vas peptide dose was expressed as the percentage ofanimals with
deferens preparation, as described above. a positive bar test. A group of eight rats was used for each
Intracerebroventricular (i.c.v.) Injections. Male Wistar rats peptide dose. The CD50 of each peptide was defined as the
weighing 220-230 g when they were received from the dose that produced a positive bar test in 50% of the animals
supplier (Charles River Breeding Laboratories) and 250-270 tested. To confirm the opioid origin of the catalepsy, nalox-
Table 1. Affinities for opioid receptors and biological activities on guinea pig ileum (GPI) and mouse vas deferens (MVD) of
dermorphin-like peptides
;t system 8 system K system GPI MVD
Peptide K;, nM OH Ki, nM 71H K;, nM T1H IC5o, nM ICso, nM
1. Tyr-DAla-Phe-Gly-Tyr-Pro-Ser-NH2 0.540 ± 0.021 0.96 929 ± 41 0.93 8162 ± 979 1.00 1.29 ± 0.11 16.5 ± 1.3
2. Tyr-DAla-Phe-Gly-Tyr-Pro-Lys-NH2 0.090 ± 0.008 0.62 1105 ± 185 617 ± 66 0.87 1.15 ± 0.13 13.6 ± 1.5
High-affinity site 0.007 ± 0.001
Low-affinity site 0.250 ± 0.013
3. Tyr-DAla-Phe-Gly-Tyr-Pro-Lys-OH 5.700 ± 0.510 0.98 1150 ± 172 0.81 - 3.82 ± 0.45 56.3 ± 7.8
4. Tyr-DAla-Phe-Gly-Tyr-Pro-Arg-NH2 0.200 ± 0.012 0.90 391 ± 25 0.98 192 ± 19 0.85 0.90 ± 0.17 10.2 ± 1.7
5. Tyr-DAla-Phe-Gly-Tyr-Pro-Asn-NH2 0.444 ± 0.031 0.89 421 ± 42 0.82 5348 ± 1175 0.86 1.57 ± 0.23 17.0 ± 2.6
6. Tyr-DAla-Phe-Trp-Asn-NH2 0.900 ± 0.052 0.85 480 ± 45 0.98 177 ± 12 1.01 5.00 ± 0.52 73.7 ± 9.1
7. Tyr-DAla-Phe-Trp-Asn-OH 4.440 ± 0.395 0.97 715 ± 73 1.01 2396 ± 351 0.89 13.10 + 1.20 205.2 ± 25.0
8. Tyr-DAla-Phe-Trp-Tyr-Pro-Asn-NH2 0.320 ± 0.026 0.95 690 ± 57 1.00 427 ± 47 1.03 0.58 ± 0.06 6.6 ± 0.9
9. Tyr-DAla-Phe-Trp-Tyr-Pro-Asn-OH 2.900 ± 0.208 0.84 865 ± 69 0.91 905 ± 61 0.97 1.30 + 0.20 10.4 + 1.3
10. Tyr-DAla-Phe-Trp-Tyr-Pro-Ser-NH2 0.390 ± 0.013 0.93 - 1.00 ± 0.15 8.7 ± 1.2
K;, equilibrium dissociation constant of the competing ligand; nH. slope of log-logit plot; IC5o, agonist concentration that produced 50%
inhibition of the electrically evoked twitch.
Pharmacology: Negri et al. Proc. Natl. Acad. Sci. USA 89 (1992) 7205

.030 F sites of different affinity, [3H]DAGO binding was displaced
by increasing concentrations of [Lys7-NH2]dermorphin (pep-
tide 5 in Table 1) in the presence of [Trp4,Asn7-
c NH2]dermorphin at three different concentrations (0.1, 0.3,
.020 I
C and 0.5 nM). For each concentration of [Trp4,Asn7-
NH2]dermorphin, Fig. 2 depicts the displacement curve fitted
.010 F by the nonlinear regression program with a two-site model.
The three competition curves show that the density of the
low-affinity site was progressively reduced as the concentra-
.005 F tion of [Trp4,Asn7-NH2]dermorphin increased from 0.1 nM to
m 0.5 nM. The peptides with a free carboxyl-terminal carboxyl
group had a two orders of magnitude lower affinity for , sites
-13 -12 -11 -10 -9 -8 -7 than their amidated analogs. Because all peptides tested had
very low affinity for 8 and K sites, they behaved as highly
[Lys7-NH2]dermorphin (log M) selective ligands for 1 sites (Table 1). Nevertheless, the new
peptides showed higher affinity for K opioid receptors of
FIG. 1. [Lys7]Dermorphin competition curve in the ,u opioid guinea pig brain than did dermorphin.
system of the rat brain. Data were fitted by the nonlinear regression Peptide Degradation. The peptides were degraded by crude
program LIGAND. The solid line was the best fit for the two-site brain homogenates (Fig. 3). An unexpected finding was that
model, the broken line that for the one-site model. The apparent within few minutes of incubation with brain homogenates the
improvement in fit for the two-site model was significant (P < 0.01
versus one-site). biological activity of [Trp4,Asn7-NH2]dermorphin and to a
lesser extent that of [Trp4]dermorphin was completely de-
one (1 mg/kg of body weight) was administered to five-rat stroyed. Since both dermorphin and Tyr-DAla-Phe-Trp-Asn-
groups 15 min before i.c.v. injection of the peptides. NH2 were degraded slowly, the rapid cleavage of [Trp4,Asn7-
NH2]dermorphin and [Trp4]dermorphin probably occurred
between Trp4 and Tyr5. The mixture of inhibitors we used did
RESULTS not afford protection against this type of enzyme degradation.
Binding Assay. Table 1 shows the results of the inhibition Biological Activity on Isolated Preparations. The peptides
of binding of [3H]DAGO, [3H][DAWa2]deltorphin I, and [3H]U- tested all acted as potent ,u opioid agonists on isolated organ
69,593 to brain membrane preparations by the peptides. preparations (Table 1). They were several times more active
With the exception of [Lys7-NH2]dermorphin (peptide 2 in in inhibiting electrically evoked contractions in guinea pig
Table 1), the peptides produced smooth competition curves ileum than in mouse vas deferens. The rank order of their
that were fitted best by a one-site model. Log-logit transfor- biological potencies roughly paralleled that of the respective
mation of [Lys7-NH2]dermorphin inhibition of the specific ,u binding affinities. However, there were notable exceptions.
[3H]DAGO binding gave a slope significantly lower than 1 [Trp4,Asn7-NH2]dermorphin had twice the potency of [Lys7-
(0.65) and an apparent K, of 0.09 ± 0.008 nM. Because the NH2]dermorphin on guinea pig ileum, but one-third the
Hofstee plot (13) was curvilinear and the line of best fit affinity for ;L opioid receptors. [Trp4,Asn7]dermorphin had
assuming a one-site model was inadequate, a nonlinear the same biological potency as [Lys7-NH2]dermorphin but a
regression analysis with a two-site model was used. This 1/20th the affinity for ,u opioid receptors.
resulted in a shallow competition curve (Fig. 1), that fitted Antinociception and Catalepsy. Peptides with a carboxyl-
data significantly better than the monophasic curve (P < 0.01 terminal amide group produced dose-related antinociceptive
versus one-site model). The computed K, values were 0.007 effects in rats (Table 2), whereas their analogs with a free
+ 0.001 nM for the high-affinity site and 0.250 + 0.013 nM for a-carboxyl group were about two orders of magnitude less
the low-affinity site. However, for this analysis we assumed active (results not shown). By i.c.v. administration, [Lys7-
that the Kd values of [3H]DAGO for the two sites were NH2]dermorphin, [Arg7-NH2]dermorphin, and [Asn7-
identical (0.78 nM). If this was not the case, the computed K, NH2]dermorphin behaved as potent analgesic agents in the
values of the competing ligand were offset by an indetermi- tail-flick test. The antinociception of [Lys7-NH2]dermorphin
nate amount. To provide further evidence for two ,u binding was long lasting (3-4 hr), exceeding that of dermorphin itself.
In contrast, [Trp4,Asn7-NH2]dermorphin, and [Trp4]dermor-
.030 phin were about 1/80th and 1/20th as active as [Lys7-NH2]-
dermorphin, respectively. Their antinociceptive effects
<.020X> 80

.010- s 0
-13 -12 -11 -10 -9 -8 -7
Time (min)
[Lys7-NHJdermorphin (logM)
FIG. 3. Time course of peptide degradation by rat brain homoge-
FIG. 2. Displacement of [3H]DAGO binding to ,u opioid receptors nates. o, [Lys7]Dermorphin (peptide 3 in Table 1); *,
by increasing concentrations of [Lys7-NH2]dermorphin in the ab- [Trp4,Asn5]dermorphin-(1-5) (peptide 7); o, dermorphin (peptide 1);
sence (o) and in the presence of [Trp4,Asn7-NH2]dermorphin (o, 0.1 A, [Trp4]dermorphin (peptide 10); and A, [Trp4,Asn7]dermorphin
nM; A, 0.3 nM; and A, 0.5 nM). (peptide 9).
7206 Pharmacology:' Negri et al. Proc. NatL. Acad. Sci. USA 89 (1992)
Table 2. Antinociceptive and cataleptic effects of dermorphin-like peptides
Peptide No. in Table 1 Mr AD50 CD50 AD5o/CD50
[Lys7-NH2]Dermorphin 2 845 0.026 0.009
± 0.364 0.126
± 0.07
[Arg7-NH2]Dermorphin 4 872 0.034 ± 0.008 0.149 ± 0.016 0.23
Dermorphin 1 804 0.035 ± 0.010 0.112 ± 0.037 0.31
[Asn7-NH2]Dermorphin 8 830 0.034 ± 0.009 0.082 ± 0.024 0.41
[Trp4,Asn5-NH2]Dermorphin-(1-5) 6 700 0.430 ± 0.044 0.210 ± 0.033 2.05
[Trp43Dermorphin 10 932 0.791 ± 0.048 0.130 ± 0.021 6.12
[Trp4,Asn7-NH2]Dermorphin 8 959 2.086 ± 0.311 0.052 ± 0.013 40.11
AD5o, ED50 (nmol per rat, i.c.v.) for antinociception in the tail-flick test; CD50, ED50 (nmol per rat, i.c.v.) for catalepsy
in the bar test.
lasted less than 1 hr. In addition to antinociception, i.c.v. affinity state of the receptor rather than a distinct ,u receptor
injection of these peptides produced catalepsy (Table 2). subtype. Our data are difficult to reconcile with this inter-
With the exception of the tryptophan-containing analogs, the pretation. The self-competition curves of DAGO were
peptides tested showed CD50 values higher than the respec- smooth and symmetrical, whereas the heterologous compe-
tive AD50 values. However, the ratio AD50/CD50 varied tition curves with [Lys7-NH2]dermorphin clearly showed two
widely among the different agonists. [Lys7-NH2]Dermorphin binding sites. In addition, displacement of [3H]DAGO bind-
showed the lowest ratio (0.073): at i.c.v. doses lower than 100 ing by [Lys7-NH2]dermorphin in the presence of [Trp4,Asn7-
pmol per rat the peptide produced antinociception alone. In NH2]dermorphin at three different concentrations (0.1, 0.3,
contrast, when given at i.c.v. doses that did not modify the and 0.5 nM) produced three biphasic competition curves in
reaction time in the tail-flick test, [Trp4,Asn7-NH2]der- which the density of [3H]DAGO binding to the low-affinity
morphin induced catalepsy. For the latter peptide we calcu- site was progressively shifted toward lower Bm.. values. A
lated AD5o/CD5o = 40, a value 500 times higher than that simple explanation for all these results might be that the two
calculated for [Lys7-NH2]dermorphin. Both antinociception binding sites represent two distinct receptor subtypes of the
and catalepsy were reversed by naloxone (data not shown). ,u system, one being the site preferred by [Lys7-NH2]-
When injected intraperitoneally (i.p.) in rats, [Lys7-NH2]- dermorphin and the other the site preferred by [Trp4,Asn7-
dermorphin produced a long-lasting antinociception that was NH2ldermorphin. Evidence also exists for distinct functional
dose-related (Fig. 4). In the range of doses tested, it did not roles of the two ,u receptor subtypes. When [Lys7-NH2]-
induce catalepsy. By i.p. administration, the AD50 of [Lys7- dermorphin was injected i.c.v. in the rats, its AD50 in the test
NH2]dermorphin (36 + 9 nmol/kg of body weight) was about of antinociception was 1/10th of its CD50 in the test of
1/40th and 1/6th that of morphine (1500 ± 260 nmol/kg) and catalepsy, whereas in the case of [Trp4,Asn7-NH2]dermor-
dermorphin (225 ± 38 nmol/kg), respectively. phin the ADso was 40 times higher than the CD50. However,
until cloning and sequence analysis provide conclusive proof,
DISCUSSION the evidence provided here for the existence of two subtypes
of the 1 receptor must be regarded with circumspection. Our
The results obtained clearly demonstrate that the dermor- experiments with brain homogenates demonstrated a large
phin-related peptides from the skin of Phyllomedusa bicolor difference in the degradation rates of the tested peptides and
and their amidated analogs are highly potent and selective u no suitable protective cocktails could be devised. The pos-
opioid agonists. In binding to opioid receptors, [Lys7- sibility that in vivo enzyme degradation may generate cleav-
NH2]dermorphin showed an affinity for the preferred u site age products that can account in part for the difference
at least 3 or 4 orders of magnitude greater than its affinity for observed in the ED50 values between the test of antinocice-
the K and 8 sites. Moreover, its competition curve for ption and that of catalepsy is not suggested by our data, but
[3H]DAGO binding was fitted best by a two-site model. At it cannot be rigorously excluded. However, both antinoci-
very low concentrations, in the picomolar range, this peptide ception and catalepsy showed classical opioid behavior as
reduced binding by some small amount, to produce a distinct evidenced by naloxone antagonism. Of the peptides tested,
change in the curve slope, followed at higher concentrations
(nanomolar) by a steeper reduction in the remaining bound [Lys7-NH21dermorphin was the most potent ju opioid agonist
[3H]DAGO. Previous data in the literature (14) show that even when injected peripherally. When given i.p., its AD50 in
DAGO binding to ,A opioid receptors sometimes reveals high- the tail-flick test was 36 nmol/kg of body weight, which is a
and low-affinity sites. The current explanation of this phe- value 1/40th of the AD50 of morphine (1500 nmol/kg). The
nomenon is that the high-affinity site represents a high- duration of the antinociceptive effect also exceeded that of
morphine. Thus its higher affinity and selectivity for ,u sites
100- and its greater antinociceptive potency, as compared with
morphine, might well make [Lys7-NH2]dermorphin a supe-
rior analgesic agent.
50- This work was supported by grants from the Italian National
Research Council (Progetto finalizzato Cimica Fine II) and the Italian
Ministry of University and of Scientific and Technological Research.

n ; ::: 1. Montecucchi, P. C., de Castiglione, R., Piani, S., Gozzini, L.
0 50 100 150 200 & Erspamer, V. (1981) Int. J. Pept. Protein Res. 17, 275-279.
2. Gyang, E. A. & Kosterlitz, H. W. (1966) Br. J. Pharmacol.
TIME (min) Chemother. 27, 514-527.
3. Basso, N., Marcelli, M., Ginaldi, A. & De Marco, N. (1985)
FIG. 4. Time course of the antinociception (analgesia) produced Peptides 6, 177-179.
by i.p. injections of [Lys7-NH2]dermorphin (o, 0.2 Amol/kg of body 4. Sandrini, G., Degli Uberti, E., Salvadori, S., Margutti, A.,
weight; o, 0.5 ,umol/kg) and morphine-hydrochloride (U, 5 ,Lmol/kg; Trasforini, G., Tomatis, R., Nappi, G. & Pansini, R. (1986)
o, 13 ,umol/kg). Bars indicate SEM. Brain Res. 371, 364-367.
Pharmacology: Negri et al. Proc. Nadl. Acad. Sci. USA 89 (1992) 7207
5. Richter, K., Egger, R., Negri, L., Corsi, R., Severini, C. 10. Portoghese, P. S., Sultana, M. & Takemori, A. E. (1988) Eur.
& Kreil, G. (1990) Proc. Natl. Acad. Sci. USA 87, 4836- J. Pharmacol. 146, 185-186.
4839. 11. Kosterlitz, H. W., Lord, J. A. H., Paterson, S. J. & Water-
6. Gillan, M. G. C., Kosterlitz, H. W. & Paterson, S. J. (1980) Br. field, A. A. (1980) Br. J. Pharmacol. 68, 333-338.
J. Pharmacol. 70, 481-490. 12. D'Amour, F. E. & Smith, D. L. (1941) J. Pharmacol. Exp.
7. Erspamer, V., Melchiorri, P., Falconieri Erspamer, G., Negri, Ther. 72, 74-79.
L., Corsi, R., Severini, C., Barra, D., Simmaco, M. & Kreil, G. 13. Molinoff, P. B., Wolfe, B. B. & Weiland, G. A. (1981) Life Sci.
(1989) Proc. Natl. Acad. Sci. USA 86, 5188-5192. 29, 427-438.
8. Goldstein, A. & Barrett, R. W. (1987) Mol. Pharmacol. 31, 14. Goldstain, A. & Naidu, A. (1989) Mol. Pharmacol. 36,265-272.
603-609. 15. Mignogna, G., Severini, C., Simmaco, M., Negri, L., Falconieri
9. Munson, P. J. & Rodbard, D. (1980) Anal. Biochem. 107, Erspamer, G., Kreil, G. & Barra, D. (1992) FEBS Lett. 302,
220-239. 151-154.